*6.3. Scenario-Specific Indicators: Leaking Cubitainer*

Another specific risk at the plant under investigation is the worn drain valves of cubitainers, leading to spills of liquid content (Figure 6). Products on floors lead to a longer breaking distance and a higher chance of losing control over the pallet mover. Figure 7 presents the bow-tie of the pre-event accident scenario of damaged loads, and more specific leaking cubitainers, complemented with possible indicators for this specific scenario. Again, only a selection of possible safety barriers and management delivery systems has been included in the bow-tie, namely, the non-technical safety barriers 'purchasing cubitainers fit for the job', 'removal of leaking cubitainers', and 'removal of leaked product', the technical safety barrier 'sandblasting floor', and the management delivery systems 'training pallet mover operators' and 'traffic management'.

**Figure 6.** A cubitainer with leaking drain valves leading to a spill of the liquid content.

**Figure 7.** The possible indicators for the pre-scenario of 'leaking cubitainer'.

#### *6.4. Evaluation of the Indicators*

Once the indicators have been determined, they have to be evaluated. Based on the specificity of the indicator and the needs of the company, this evaluation can take place on a yearly basis, or if needed with a lower or a higher frequency.

Additionally, after the development of the indicators, targets and limits have to be assigned to every indicator. This means that the company has to decide what is acceptable as a result and what is not. In the example of the coverage ratio of the training, a target could be that 100% of all pallet mover operators should be trained for pallet mover use. In the example of the number of leaking cubitainers, a target could be that <5% of the cubitainers is leaking.

An important aspect is that responsibilities have to be indicated: Who does what and when in order to reach the goals of the indicators. In the example of the (re)training, responsibility has to be indicated for the subscriptions for the (re)training. The same applies to responsibilities to take actions when a target is not achieved.

Based on the elaborated examples, it could seem that a lot of collection and registration is needed for all indicators. However, it should be noted that the given examples and their accompanying indicators are very specific. Once the entire set of indicators has been developed, it will become clear that many of them can be collected and registered under the same heading.

In order to facilitate an adequate monitoring of the indicators, a system should be set up to report and to collect the required data. Such systems are often already (partly) present in a company.

To be complete, something should be said on the necessity of the indicators, which is a reflection that should be made before developing the indicators. After all, it should first be analyzed if the processes that are present in the company are as inherently safe as possible [41]. In the given example regarding the narrow maneuvering spaces, whether a pallet mover is the safest equipment to use at these places (and by extension at all places) should be considered. In the example of the leaking cubitainers, whether cubitainers are necessary and if the process cannot be designed in such a way that the transportation of the liquid is minimalized (and by extension that a large amount of manipulation of all loads at the plant is minimalized) should be considered. Kletz's [42] article entitled "what you don't have, can't leak" is a good resume of this matter.

#### **7. Discussion and Conclusions**

The outcomes deriving from the current accident analysis that is used at the Belgian plant under investigation (6W-2H and why-why technique) does not seem sufficient to take adequate measures in order to prevent accidents with pallet movers. When preventive measures are taken based on accident analyses, several shortcomings can be identified. Firstly, preventive measures based on accident analyzes only focus on accidents that already occurred, leaving all other potential accident scenarios out of scope. A method that includes not only company specific data, but also generic data sources such as the literature and national accident data, generates information on the entire accident process, including aspects that have not (yet) occurred at a specific plant. Additionally, it can be concluded that the recommendations resulting from the current accident analysis that is used at the plant under investigation, are mostly individual-oriented. This is however not an intrinsic problem of the 6W-2H and why-why technique, as this technique focusses on both technical, organizational as human aspects. Hence, the focus on the human aspects is not a consequence of the technique itself, but of the way the technique is applied.

To address the shortcomings of the current accident analysis, an alternative method—i.e., the bow-tie method—is chosen in order to address pallet mover safety. The bow-tie was chosen for several reasons. Firstly, the bow-ties were composed using a multi-method design. This multi-method design leads to a better comprehensiveness of the entire accident process of pallet mover use and gives a detailed overview of what could possibly go wrong with a pallet mover. In the bow-ties, the possible causes and consequences of potential accidents are identified. Additionally, the bow-tie includes the influence of safety measures (safety barriers and management delivery systems) on the evolution of accident scenarios [43].

Due to the use of a multi-method design to compose the bow-ties, not only company specific data were included, but also generic data sources such as the literature and national accident data. This leads to information on the entire accident process, including aspects that have not (yet) occurred at a specific plant.

Because of the comprehensive character of the bow-tie method, the results are easily transferable to other production facilities where pallet movers are used for internal transport, assuming that the hazards are the same. This means that, if this study was conducted in another production facility with similar hazards and similar a working environment, composition of the bow-ties would have led to a similar outcome as in Tables 1 and 2 (this does not mean that the process of linking indicators to the bow-ties is the same, as this is very company specific). In other words, the bow-tie method leads to a general model that is transferable and applicable in every setting where, in this case, pallet movers are being used. However, the indicators may be different.

Another advantage of the bow-tie method is that it allows us to make a clear distinction between preventing and mitigating safety barriers and management delivery systems.

Seven hazards regarding pallet mover use could be identified based on the composition of the bow-ties: Load, the speed of the pallet mover, acceleration of the pallet mover, the design of the workplace, conditions of the workplace, conditions of materials (load and pallet mover), and operating the pallet mover. Through several identified pre-event accident scenarios, these hazards can lead to different central events: Instability of the load, loss of control over the pallet mover, and a breakdown of the pallet mover. At their turn, these central events can lead through several post-accident scenarios to different consequences: Injury, damage, or economic loss. Several technical and non-technical safety barriers and management delivery systems to prevent or mitigate the central event could be linked to the accident scenarios.

The identified safety barriers and management delivery systems mainly focus on organizational aspects, and, to a lesser extent, on the individual behavioral aspects of operators. The pitfall of 'blaming the victim', which is often present in other methods of accident analysis where there is primarily focused on the individual behavior of the operators, is therefore reduced when using the bow-tie method.

Once bow-ties are composed and safety barriers and management delivery systems have been identified, indicators should be developed and monitored consequently. These indicators should be composed based on their applicability in the company, meaning what is possible given a specific company environment. When developing indicators, an important distinction can be made. Firstly, there are general indicators. In the bow-ties, certain management delivery systems can be linked to many of the accident scenarios. When indicators are developed for frequently occurring management delivery systems, these indicators can be considered as general because they are not linked to only one scenario. Secondly, there are scenario-specific indicators. This means that indicators are linked to specific scenarios that require attention in a plant. With both the general and the scenario-specific indicators, a certain sequentiality should be acknowledged in the follow-up of the indicators. For all indicators, it is therefore important to set priorities.

To conclude, indicators are an important result of a bow-tie analysis. When a company reaches a consensus on a set of indicators to be monitored, a unique insight is obtained on the status and development of potential accident scenarios. Management can intervene adequately to ensure a safe production.

**Author Contributions:** Conceptualization, K.v.N., P.S. and G.R.; Methodology, K.v.N., P.S. and G.R.; Validation, K.v.N., P.S. and G.R.; Formal Analysis, K.v.N and P.S.; Investigation, K.v.N and P.S.; Data Curation, K.v.N., P.S. and G.R.; Writing-Original Draft Preparation, K.v.N.; Writing-Review & Editing, P.S., G.R., N.P., O.A. and K.P.; Visualization, K.v.N.; Supervision, G.R.; Project Administration, K.v.N.; Funding Acquisition, G.R."

**Acknowledgments:** The authors would like to thank the company under investigation in this article for their cooperation and for the possibility to conduct this research.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Appendix A**

**Table A1.** An example of a recordable accident with a pallet mover at the Belgian plant (input based on the 6W-2H and why-why analysis).
